giovanni francesco ricci , marco centanni , francesco
TRANSCRIPT
Simulating streamflow in a temporary karst river system
Giovanni Francesco Ricci1 , Marco Centanni1, Francesco Gentile1, Anna Maria De Girolamo2
(1) University of Bari, Department of Agricultural and Environmental Sciences, Bari, Italy(2) Water Research Institute-National Research Council, Bari, Italy
Thursday, 29 AprilSession HS2.1.3 Zero flow: hydrology and biogeochemistry of intermittent and ephemeral streams
1 •Background and Aims
2 •Study Area
3 •Materials and methods
4 •Ongoing results
5 •Conclusions
Background and aims
Factor influencing flow regime:
• Spatial gradient in rainfall and temperature• Lithology and geological features (i.e. limestone, sinkhole)• Anthropic pressure (i.e. water treatment plants, water abstractions)
Most of the basins in the Mediterranean Region cease flow for some time of the year, hence are defined as non-perennial rivers (i.e. temporary rivers, intermittent rivers, ephemeral streams, episodic streams).
Background and aims
Aims:
• Set up SWAT model in a karst temporary Mediterranean basin characterized by paucity of data
• Define problem-solving in simulating hydrology with a particular attention to zero flow periods
Problems:
• Data of land use, soil and slope not homogeneous• limited time series of streamflow due to the reduced
investments in monitoring intermittent river • Climate stations not well distributed in the basin • Presence of several gaps in weather time series
In such a complex environment, the hydrological and water quality model set up and run may be challenging.
Study area
The Canale D'Aiedda (Puglia, Italy) is a temporary karst river basin
Morphological characteristics Canale d’Aiedda basin Nil wadi Basin
Drainage area 360 km2 304.22 km2
Main channel length (km) 29 km 35 km
Mean elevation (m a.s.l.) 131 m (a.s.l.) 451 m (a.s.l.)
Maximum elevation (m a.s.l.) 381 m (a.s.l.) 1510 m (a.s.l.)
Minimum elevation (m a.s.l.) 0 m (a.s.l.) 0 m (a.s.l.)
Mean watershed slope (%) 4.70% 26.95%
Mean slope of the main channel (%) 0.84% 2.98%
Study area – monitoring equipment
Study area
Main Land uses of the basinAreas with outcropping limestone (purple) and areas with clay soils
Methodology
The Soil and Water Assessment Tool (SWAT) considering two different set up solution:
1. All the basin area was considered2. The area with outcropping limestone was cut off with a mask
Land Cover Map
Soil Map
Calibration
SWAT Model Set up
Observed Discharge
Digital Elevation
Model (DEM)
1
2
Data Source Resolution
Digital Terrain Model Puglia Region (http://www.sit.puglia.it) 8 × 8 mLand use map Puglia Region (http://www.sit.puglia.it) 1:5000
National Agricultural Census (http://censimentoagricoltura.istat.it/index.php?id=73) –Soil map and database Puglia Region (2001) 1:100000
JRC-ESDAC (https://esdac.jrc.ec.europa.eu/resource-type/datasets) 500 × 500 m
Point sources Apulian Water Authority (Personal communication) (W1, W2, W3)
–Puglia Region (http://www.sit.puglia.it) (S1, S2)Regional Agency for Environmental Protection (http://www.arpa.puglia.it/web/guest/depuratori) (W1, W2, W3)Sampling and chemical analysis of treated effluents (W2, W3)
Meteorological data Civil Protection Service - Puglia Region (https://protezionecivile.puglia.it/)Daily 1/1/1997 -
31/12/2019Regional Agency for Irrigation and Forestry Activities (http://www.agrometeopuglia.it/)National Agency for New Technologies, Energy and Sustainable Economic Development (http://www.solaritaly.enea.it/)
Gauging station Department of Agricultural and Environmental Sciences, University of Bari Aldo MoroDaily 8/2017 -
12/2019
Agricultural practices Interviews with farmers and agricultural advisors (D'Ambrosio et al., 2019). –
Methodology – data used to set up and calibrate the model
Ongoing results
Infiltration and transmission losses and soil hydraulic parameters resulted the most relevant in simulating hydrology
Configuration 1 Configuration 2
R2: 0.43NSE: 0.39PBIAS: +26.9
R2: 0.72NSE: 0.71PBIAS: +5.1
Ongoing results
Configuration 1
Configuration 2
Both simulations (1 and 2) show an asymptotic overestimation in low flow periods when the lowest measured values was 0.008 m3 s-1
Extreme low flows are affected by large uncertainty.
The comparison of measured and modeled flow duration curves and the analysis of pictures of several reaches of the Canale d’Aiedda took in field surveys, lead to identify a “zero flow” threshold used to rectify the simulated streamflow.
Qmin = 0.010 m3 s-1
Qmin = 0.014 m3 s-1
Qmin = 0.012 m3 s-1
Qmin = 0.002 m3 s-1
Ongoing results
The analysis of the pictures taken in different time period and in different location upstream to the gauging station evidenced that some reaches were intermittent during the year
• The SWAT model is able to simulate daily streamflow with good performances especiallyconsidering the configuration without outcropping limestone.
• Unsatisfactory results were obtained for when all the basin area was considered,highlighting that SWAT needs to be further calibrated parameterizing transmissionlosses and soil hydraulic properties
• Streamflow is affected by uncertainty which lead to overestimate the baseflow
• The hydrograph of calibrated model in upstream river sections does not show anabsence of flow in the dry season.
• A procedure was defined to improve the no-flow simulations. A «zero-flow» threshold(actual no flow) was identified for each intermittent reaches through field surveys.
• The simulated streamflow time series was rectified for the dry season based on zero-flow thresholds .
Conclusions
Any question? Join the chat session
Thursday, 29 AprilSession HS2.1.3 Zero flow: hydrology and biogeochemistry of intermittent and ephemeral streams